Hindawi Publishing Corporation Journal of Inequalities and Applications Volume 2007, Article ID 86052, 22 pages doi:10.1155/2007/86052 Research Article Some Relationships between the Analogs of Euler Numbers and Polynomials C. S. Ryoo, T. Kim, and Lee-Chae Jang Received 5 June 2007; Revised 28 July 2007; Accepted 26 August 2007 Recommended by Narendra K. Govil We construct new twisted Euler polynomials and numbers. We also study the generating functions of the twisted Euler numbers and polynomials associated with their interpola- tion functions. Next we construct twisted Euler zeta function, twisted Hurwitz zeta func- tion, twisted Dirichlet l-Euler numbers and twisted Euler polynomials at non-positive integers, respectively. Furthermore, we find distribution relations of generalized twisted Euler numbers and polynomials. By numerical experiments, we demonstrate a remark- ably regular structure of the complex roots of the twisted q-Euler polynomials. Finally, we give a table for the solutions of the twisted q-Euler polynomials. Copyright © 2007 C. S. Ryoo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1. Introduction and notations Throughout this paper, we use the following notations. By Z p we denote the ring of p-adic rational integers, Q denotes the field of rational numbers, Q p denotes the field of p-adic rational numbers, C denotes the complex numbers field, and C p denotes the completion of algebraic closure of Q p .Letν p be the normalized exponential valuation of C p with |p| p = p −ν p (p) = p −1 . When one talks of q-extension, q is considered in many ways such as an indeterminate, a complex number q ∈ C,orp-adic number q ∈ C p .Ifq ∈ C,one normally assumes that |q| < 1. If q ∈ C p , we normally assume that |q − 1| p <p −1/(p−1) so that q x = exp(xlogq), for |x| p ≤ 1. [x] q = [x : q] = 1 − q x 1 − q cf. [1–18] . (1.1) 2 Journal of Inequalities and Applications Hence, lim q→1 [x] = x for any x with |x| p ≤ 1inthepresentp-adic case. Let d be a fixed integer and let p be a fixed prime number. For any positive integer N,weset X = lim ←− N Z dp N Z , X ∗ = 0<a<dp (a,p)=1 (a + dpZ p ), a + dp N Z p = x ∈ X | x ≡ a moddp N , (1.2) where a ∈ Z lies in 0 ≤ a<dp N . For any positive integer N, μ q a + dp N Z p = q a dp N q (1.3) is known to be a distribution on X (cf. [1–18]). For g ∈ UD Z p = g | g : Z p → C p is a uniformly differentiable function , (1.4) the p-adic q-integral was defined by [1, 2, 6–18] I q (g) = Z p g(x)dμ q (x) = lim N→∞ 1 p N 0≤x<p N g(x)q x . (1.5) Note that I 1 (g) = lim q→1 I q (g) = Z p g(x)dμ 1 (x) = lim N→∞ 1 p N 0≤x<p N g(x) (1.6) (see [1, 2, 6–18]). For q ∈ [0,1], certain q-deformed bosonic operators may be introduced which generalize the undeformed bosonic ones (corresponding q = 1); see [1, 2, 6–18]. For g ∈ UD(Z p ) Z p g(x)dμ 1 (x) = X g(x)dμ 1 (x) (1.7) (see [6–18] for details). We assume that q ∈ C with |1 − q| p < 1. Using definition, we note that I 1 (g 1 ) = I 1 (g)+ g (x), where g 1 (x) = g(x +1). Let T p = m≥1 C p m = lim m→∞ C p m , (1.8) where C p m ={w | w p m = 1} is the cyclic group of order p m .Forw ∈ T p , we denote by φ w : Z p → C p the locally constant function x −→ w x .Ifwetake f (x) = φ w (x)e tx ,thenwe C. S. Ryoo et al. 3 easily see that Z p φ w (x)e tx dμ 1 (x) = t we t − 1 . (1.9) Kim [8] treated the analog of Bernoulli numbers, which is called twisted Bernoulli num- bers. We define the twisted Bernoulli polynomials B n,w (x) e xt t we t − 1 = ∞ n=0 B n,w (x) t n n! . (1.10) Using Taylor series of e tx in the above equation, we obtain Z p x n φ w (x)dμ 1 (x) = B n,w , (1.11) where B n,w = B n,w (0). The Euler numbers E n are usually defined by means of the following generating func- tion: e Et = 2 e t +1 = ∞ n=0 E n t n n! cf. [1–18] , (1.12) where the symbol E n is interpreted to mean that E n must be replaced by E n when we expand the one on the left. These numbers are classical and important in mathematics and in various places like analysis, number theory. Frobenius extended such numbers as E n to the so-called Frobenius-Euler numbers H n (u) belonging to an algebraic number u with |u| > 1. Let u be an algebraic number. For u ∈ C with |u| > 1, the Frobenius-Euler numbers H n (u)belongingtou are defined by the generating function e H(u)t = 1 − u e t − u = ∞ n=0 H n (u) t n n! cf. [6–10] , (1.13) with the usual convention of symbolically replacing H n by H n . The Euler polynomials E n (x)aredefinedby e E(x)t = 2 e t +1 e xt = ∞ n=0 E n (x) t n n! cf. [6–16] . (1.14) For u ∈ C with |u| > 1, the Frobenius-Euler polynomials H n (u,x)belongingtou are de- fined by e H(u,x)t = 1 − u e t − u e xt = ∞ n=0 H n (u,x) t n n! cf. [6–18] . (1.15) Kim gave a relation between B n,w and H n (u), with nth Euler numbers as follows: B n,w = n w − 1 H n−1 (w −1 ), if w = 1. (1.16) 4 Journal of Inequalities and Applications Now, we consider the case q ∈ (−1,0) corresponding to q-deformed fermionic certain and annihilation operators and the literature given therein [6–18]. The expression for the I q (g) remains the same, so it is tempting to consider the limit q →−1. That is, I −1 (g) = lim q→−1 I q (g) = Z p g(x)dμ −1 (x) = lim N→∞ 0≤x<p N g(x)(−1) x . (1.17) Let g 1 (x) be translation with g 1 (x) = g(x + 1). Then we see that I −1 g 1 =− lim N→∞ p N −1 x=0 g(x)(−1) x +2g(0) =−I −1 ( f )+2g(0). (1.18) Therefore, we obtain the following lemma. Lemma 1.1. For g ∈ UD(Z p ), one has I −1 g 1 + I −1 (g) = 2g(0). (1.19) From (1.19), we can easily derive the following theorem. Theorem 1.2. For g ∈ UD(Z p ), n ∈ N, one has I −1 g n = (−1) n I −1 (g)+2 n−1 l=0 (−1) n−1−l g(l), (1.20) where g n (x) = g(x + n). Corollar y 1.3. For g ∈ UD(Z p ), n(= odd) ∈ N, one has I −1 g n + I −1 (g) = 2 n−1 l=0 (−1) l g(l). (1.21) By Lemma 1.1, we can consider twisted Euler numbers. If we take g(z) = φ w (z)e tz , (w ∈ T p ), then we have I −1 φ w (z)e tz = 2 we t +1 = ∞ n=0 E n,w t n n! . (1.22) Now we define twisted Euler numbers E n,w as follows: F w (t) = 2 we t +1 = ∞ n=0 E n,w t n n! . (1.23) Using Taylor series of e zt above, we obtain E n,w = Z p w z z n dμ −1 (z). (1.24) C. S. Ryoo et al. 5 For w ∈ T p , we introduce the twisted Euler polynomials E n,w (z). Twisted Euler poly- nomials E n,w (z) are defined by means of the generating function F w (t,z) = 2 we t +1 e zt = I −1 φ w (x)e t(z+x) = ∞ n=0 E n,w (z) t n n! , (1.25) where E n,w (0) = E n,w . Using Taylor series of e tx in the above equation, we have E n,w (x) = Z p φ w (x)(x + z) n dμ −1 (x) = n k=0 n k z n−k Z p φ w (x)x k dμ −1 (x). (1.26) Thus we easily see that E n,w (z) = n k=0 n k z n−k E k,w . (1.27) Let χ be the Dirichlet character with conductor f ( = odd) ∈ N.Ryooetal.[16]studied the generalized Euler numbers and polynomials. The generalized Euler numbers associ- ated with χ, E n,χ , were defined by means of the generating function F χ (t) = 2 f −1 a =0 χ(a)(−1) a e at e ft +1 = ∞ n=0 E n,χ t n n! . (1.28) Generalized Euler polynomials, E n,χ (x), were also defined by means of the generating function F χ (t,z) = 2 f −1 a =0 χ(a)(−1) a e at e ft +1 e zt = ∞ n=0 E n,χ (z) t n n! . (1.29) Substituting g(x) = χ(x)φ w (x)e tx into (1.21), then the generalized twisted Euler numbers E n,χ,w are defined by means of the generating functions F χ,w (t) = X φ w (x)e tx χ(x)dμ −1 (x) = 2 f −1 a =0 e ta (−1) a χ(a)φ w (a) φ w ( f )e ft +1 = ∞ n=0 E n,χ,w t n n! . (1.30) Using the above equation, E n,χ,w are defined by E n,χ,w = X φ w (x)x n χ(x)dμ −1 (x). (1.31) Generalized twisted Euler polynomials, E n,χ,w (z), are defined by F χ,w (t,z) = F χ,w (t)e zt = X φ w (x)e tx χ(x)dμ −1 (x)e tz = 2 f −1 a =0 e ta (−1) a χ(a)φ w (a) φ w ( f )e ft +1 e zt = ∞ n=0 E n,χ,w (z) t n n! . (1.32) 6 Journal of Inequalities and Applications We set F χ,w (t,z) = 2 f −1 a =0 (−1) a χ(a)φ w (a)e (a+z)t φ w ( f )e ft +1 . (1.33) Using the above equation, we have ∞ n=0 E n,χ,w (z) t n n! = X φ w (x)e tx χ(x)dμ −1 (x)e tz = X φ w (x)e t(x+z) χ(x)dμ −1 (x) = ∞ n=0 X φ w (x)(x + z) n χ(x)dμ −1 (x) t n n! = ∞ n=0 n k=0 n k z n−k X φ w (x)x k χ(x)dμ −1 (x) t n n! . (1.34) Using the comparing coefficients t n /n!, we easily see that E n,χ,w (z) = n k=0 n k z n−k E k,χ,w . (1.35) We have the following remark. Remark 1.4. Note that (1) if w → 1, then F χ,w (t,z) → F χ (t,z)andE n,χ,w (z) → E n,χ (z); (2) ∞ n=0 E n,χ,w (z) t n n! = 2 f −1 a =0 e ta (−1) a χ(a)φ w (a) φ w ( f )e ft +1 e zt = ∞ n=0 E n,χ,w (z) t n n! ∞ n=0 z n t n n! . (1.36) Using the Cauchy product in the right-hand side of the above equation in (2), we obtain ∞ n=0 E n,χ,w (z) t n n! = ∞ n=0 n k=0 E k,χ,w z n−k t n k!(n − k)! . (1.37) Comparing the coefficients t n on both sides of the above equation, we arrive at (1.35). 2. Twisted zeta function In this section, we introduce the twisted Euler zeta function and twisted Hurwitz-Euler zeta function. We derive a new twisted Hurwitz-type l-function which interpolates the C. S. Ryoo et al. 7 generalized Euler polynomials E n,χ,w (x). We give the relation between twisted Euler num- bers and twisted l-functions at nonpositive integers. Let χ be the Dirichlet character with conductor f ( = odd) ∈ N.Weset F χ,w (t) = 2 f −1 a =0 e ta (−1) a χ(a)φ w (a) φ w ( f )e ft +1 , − π f − logw<t< π f − log . (2.1) By (2.1), we see that F χ,w (t) = 2 ∞ m=1 χ(m)w m (−1) m e tm . (2.2) From (1.30)and(2.2), we note that d k dt k F χ,w (t)| t=0 = 2 ∞ m=1 χ(m)w m (−1) m m k (k ∈ N). (2.3) Therefore, we obtain the following theorem. Theorem 2.1. For k ∈ N, one has E k,χ,w = 2 ∞ m=1 χ(m)w m (−1) m m k (k ∈ N). (2.4) Thus we define the twisted Dirichlet-type l-series as follows. Definit ion 2.2. For s ∈ C, define the Dirichlet-type l-series related to twisted Euler num- bers, l w (s,χ) = 2 ∞ n=1 χ(n)(−1) n w n n s . (2.5) Theorem 2.3. For k ∈ N, one has l w (−k,χ) = E k,χ,w . (2.6) Next, we introduce the Hurwitz-type twisted Euler zeta function. Since F w (t,z) = 2 we t +1 e zt = ∞ n=0 E n,w (z) t n n! , (2.7) we obtain F w (t,z) = 2 ∞ n=0 (−1) n w n e (n+z)t . (2.8) From (2.8), we note that d k dt k F w (t,z)| t=0 = 2 ∞ n=0 (−1) n w n (n + z) k (k ∈ N). (2.9) 8 Journal of Inequalities and Applications Therefore, we have the following theorem. Theorem 2.4. For k ∈ N, one has E k,w (z) = 2 ∞ n=0 (−1) n w n (n + z) k (k ∈ N). (2.10) Thus the twisted Hurwitz-Euler zeta function is defined as follows. Definit ion 2.5. Let s ∈ C.Then ζ E,w (s,z) = 2 ∞ n=0 (−1) n w n (n + z) s . (2.11) By Theorem 2.4 and Definition 2.5, we have the following theorem. Theorem 2.6. For k ∈ N,oneobtains ζ E,w (−k,z) = E k,w (z). (2.12) Let us define two-variable twisted Euler numbers attached to χ as follows. By (1.33), we see that F χ,w (t,z) = 2 ∞ n=0 (−1) n χ(n)w n e (n+z)t . (2.13) From (2.13), we note that d k dt k F χ,w (t,z)| t=0 = 2 ∞ n=0 (−1) n χ(n)w n (n + z) k (k ∈ N). (2.14) Therefore, we obtain the following theorem. Theorem 2.7. For k ∈ N, one has E k,χ,w (z) = 2 ∞ n=0 (−1) n χ(n)w n (n + z) k (k ∈ N). (2.15) Hence we define two-variable twisted l-series as follows. Definit ion 2.8. For s ∈ C.Then l w (s,χ | z) = 2 ∞ n=0 (−1) n χ(n)w n (n + z) s . (2.16) Therelationbetweenl w (−k,χ | z)andE k,χ,w (z) is given by the following theorem. Theorem 2.9. For k ∈ N,oneobtains l w (−k,χ | z) = E k,χ,w (z). (2.17) C. S. Ryoo et al. 9 Note that l w (s,χ | z) = 2 ∞ n=0 (−1) n χ(n)w n (n + z) s = f −s f −1 a=0 (−1) a w a χ(a)ζ E,w f s, a + z f . (2.18) Therelationbetweenl w (s,χ | z)andζ E,w (s,z) is given by the following theorem. Theorem 2.10. For s ∈ C, l w (s,χ | z) = 1 f s f −1 a=0 (−1) a w a χ(a)ζ E,w f s, a + z f . (2.19) Observe that, substituting z = 0into(2.19), l w (s,χ) = f −s f −1 a=0 (−1) a w a χ(a)ζ E,w f s, a f . (2.20) Substituting s =−n with n ∈ N, l w (−n,χ | z) = f n f −1 a=0 (−1) a w a χ(a)ζ E,w f − n, a + z f . (2.21) By Theorem 2.6 and (2.21), we have l w (−n,χ | z) = f n f −1 a=0 (−1) a w a χ(a)E n,w f a + z f . (2.22) Using (1.27), we get l w (−n,χ | z) = f n f −1 a=0 (−1) a w a χ(a) n k=0 n k f k−n (a + z) n−k E k,w f = f n f −1 a=0 (−1) a w a χ(a) n k=0 n k f k−n n −k j=0 n − k j z n−k− j a j E k,w f = n k=0 n −k j=0 n k n − k j f k z n−k− j E k,w f S w,χ (j), (2.23) where S w,χ (j) = f −1 a=0 (−1) a w a χ(a)a j . (2.24) By Theorem 2.9 and (2.23), we arrive at the following theorem. 10 Journal of Inequalities and Applications Theorem 2.11. For n ∈ Z + ,oneobtains E n,χ,w (z) = n k=0 n −k j=0 n k n − k j f k z n−k− j E k,w f S w,χ (j). (2.25) 3. Twisted q-Euler zeta function and twisted q-analog Dir ichlet l-function Our primary goal of this section is to define generating functions of the twisted q-Euler numbers and polynomials. Using these functions, twisted q-zeta function and twisted q- l-functions are defined. These functions interpolate twisted q-Euler numbers and gener- alized twisted q-Euler numbers, respectively. Now, we introduce the generating functions F q (t)andF q (x,t). Ryoo et al. [15]treatedtheanalogofEulernumbers,whichiscalled q-Euler numbers in this paper. Using p-adic q-integral, we defined the q-Euler numbers as follows: E n,q = Z p [t] n q dμ −q (t), for n ∈ N. (3.1) Thus we obtain E n,q = [2] q 1 1 − q n n l=0 n l (−1) l 1 1+q l+1 , (3.2) where n i is the binomial coefficient (see [1–18]). Using the above equation, we have F q (t) = [2] q ∞ m=0 (−1) m q m e [m] q t . (3.3) Thus q-Euler numbers, E n,q , are defined by means of the generating function F q (t) = [2] q ∞ n=0 (−1) n q n e [n] q t . (3.4) Note that ∞ n=0 E n,q t n n! = Z p ∞ n=0 [x] n q n! t n dμ −q (x) = Z p e [x] q t dμ −q (x). (3.5) Thus we have Z p e [x] q t dμ −q (x) = [2] q ∞ n=0 (−1) n q n e [n] q t . (3.6) Similarly, the generating function F q (t,z)oftheq-Euler polynomials E n,q (z)isdefined analogously as follows: F q (t,z) = ∞ n=0 E n,q (z) t n n! = [2] q ∞ n=0 (−1) n q n e [n+z] q t . (3.7) [...]... the recent paper [10], Kim and Rim constructed (h, q)-extensions of the twisted Euler numbers and polynomials They also defined (h, q)generalizations of the twisted zeta function and L-series These numbers and polynomials are considered as the (h, q)-extensions of their previous results However, these (h, q )Euler numbers and generating functions do not seem to be natural extension of Euler numbers and. .. of the Jangjeon Mathematical Society, vol 8, no 1, pp 13–17, 2005 [8] T Kim, “q -Euler numbers and polynomials associated with p-adic q-integrals,” Journal of Nonlinear Mathematical Physics, vol 14, no 1, pp 15–27, 2007 [9] T Kim, “A note on some formulas for the q -Euler numbers and polynomials,” Proceedings of the Jangjeon Mathematical Society, vol 9, pp 227–232, 2006 [10] T Kim and S.-H Rim, “On the. .. q -Euler numbers and polynomials and the Euler numbers and polynomials which are treated in this paper In [9], many interesting integral equations related to fermionic p-adic integrals on Z p are known We proceed by first constructing generating functions of the twisted q -Euler polynomials and numbers Then, by applying Mellin transformation to these generating functions, integral representations of the. .. for q-twisted and q-generalized twisted Euler numbers, ” Advanced Studies in Contemporary Mathematics, vol 9, no 2, pp 203–216, 2004 [3] A Kudo, “A congruence of generalized Bernoulli number for the character of the first kind,” Advanced Studies in Contemporary Mathematics, vol 2, pp 1–8, 2000 [4] Q.-M Luo, Some recursion formulae and relations for Bernoulli numbers and Euler numbers of higher order,”... in the field of research of the En,q,w (x) to appear in mathematics and physics The reader may refer to [11, 14–16] for the details We calculated an approximate solution satisfying En,q,w (x), N = 2,4, q = 1/2, x ∈ C The results are given in Tables 4.2 and 4.3 5 Further remarks and observations Using p-adic q-fermionic integral, Rim and Kim [13] studied explicit p-adic expansion for alternating sums of. .. polynomials,” to appear in Journal of Computational and Applied Mathematics [15] C S Ryoo, T Kim, and R P Agarwal, “A numerical investigation of the roots of q-polynomials,” International Journal of Computer Mathematics, vol 83, no 2, pp 223–234, 2006 [16] C S Ryoo, T Kim, and L C Jang, “A note on generalized Euler numbers and polynomials,” International Journal of Computer Mathematics, vol 84, no 7, pp 1099–1111,... (s,z) is a meromorphic function on C The relation between ζq,w (s,z) and Ek,q,w (z) is given by the following theorem Theorem 3.4 For k ∈ N, one has ζq,w (−k,z) = Ek,q,w (z) (3.24) Observe that ζq,w (−k,z) function interpolates Ek,q,w (z) numbers at nonnegative integers 4 Distribution and structure of the zeros In this section, we investigate the zeros of the twisted q -Euler polynomials En,q,w (z) by using... Find the numbers of complex zeros CEn,q,w (x) of En,q,w (x), Im(x) = 0 Prove or give a counterexample 18 Journal of Inequalities and Applications 3 2 Im(z) 1 0 1 2 1 0 1 Re(z) 2 3 4 3 4 Figure 4.9 Zeros of E10,q,w (x) 3 2 Im(z) 1 0 1 2 1 0 1 Re(z) 2 Figure 4.10 Zeros of E20,q,w (x) Conjecture Since n is the degree of the polynomial En,q,w (x), the number of real zeros REn,q,w (x) lying on the real... order,” Advanced Studies in Contemporary Mathematics, vol 10, no 1, pp 63–70, 2005 [5] Q.-M Luo and F Qi, Relationships between generalized Bernoulli numbers and polynomials and generalized Euler numbers and polynomials,” Advanced Studies in Contemporary Mathematics, vol 7, no 1, pp 11–18, 2003 [6] T Kim, “q-Volkenborn integration,” Russian Journal of Mathematical Physics, vol 9, no 3, pp 288–299,... and S.-H Rim, “On the twisted q -Euler numbers and polynomials associated with basic q-l-functions,” Journal of Mathematical Analysis and Applications, vol 336, no 1, pp 738–744, 2007 22 Journal of Inequalities and Applications [11] T Kim, C S Ryoo, L C Jang, and S.-H Rim, “Exploring the q-Riemann zeta function and qBernoulli polynomials,” Discrete Dynamics in Nature and Society, vol 2005, no 2, pp . Corporation Journal of Inequalities and Applications Volume 2007, Article ID 86052, 22 pages doi:10.1155/2007/86052 Research Article Some Relationships between the Analogs of Euler Numbers and Polynomials C In the recent paper [10],KimandRimconstructed (h,q)-extensions of the twisted Euler numbers and polynomials. They also defined (h,q)- generalizations of the twisted zeta function and L-series. These. numbers and polynomials are considered as the (h,q)-extensions of their previous results. However, these (h,q)- Euler numbers and generating functions do not seem to be natural extension of Euler numbers